Component Attention Guided Face Super-Resolution Network: CAGFace

TL;DR

This paper introduces CAGFace, a multi-stage convolutional neural network that employs facial component attention maps to enhance 4x face image super-resolution, achieving superior quantitative and perceptual results.

Contribution

The novel integration of component-wise attention maps via segmentation to guide face super-resolution in a multi-stage network is the key innovation.

Findings

Achieves higher PSNR and SSIM compared to state-of-the-art methods.

Produces more perceptually pleasing face images.

Effectively utilizes intermediate reconstructions for progressive enhancement.

Abstract

To make the best use of the underlying structure of faces, the collective information through face datasets and the intermediate estimates during the upsampling process, here we introduce a fully convolutional multi-stage neural network for 4$\times$ super-resolution for face images. We implicitly impose facial component-wise attention maps using a segmentation network to allow our network to focus on face-inherent patterns. Each stage of our network is composed of a stem layer, a residual backbone, and spatial upsampling layers. We recurrently apply stages to reconstruct an intermediate image, and then reuse its space-to-depth converted versions to bootstrap and enhance image quality progressively. Our experiments show that our face super-resolution method achieves quantitatively superior and perceptually pleasing results in comparison to state of the art.